[Pw_forum] Large shift of the Fermi energy - phonons, slab

[Nicki Frank Hinsche]

Dear community,

I am puzzled with some non-converging phonon calculations and  
especially with a very large shift of the Fermi energy in the  
perturbation step.
I am doing some calculations on metallic slabs, comparable to Eiguren  
et al. Phys. Rev. Lett. (2002) vol. 88 (6) 066805 or
Hofmann et al. New Journal of Physics (2009) vol. 11 (12) 125005. It  
worked out for several systems, with various thicknesses but now I  
become
desperate calculating the phonons of a simple 21layer slab of Cu(111).  
The vac. is around 2.5nm. The electronic structure looks fine, the  
relaxation
is done - but the phonons fail for the Gamma point (only). The Fermi  
energy shift is unphysically large. The DOS has no resonance nearby.

So what could be the problem? Any hints would be helpful.

Thanks Nicki

following some input/output for the phonons on (away) the Gamma point:

Input:

&inputph
   reduce_io = .false.
   tr2_ph=1.0d-12,
   prefix='Cu',
   alpha_mix = 0.150
   ldisp=.true.,
   nq1=4, nq2=4, nq3=1
   start_q=1
   last_q=1
   recover = .true.,
   amass(1)=63.546
   outdir="***",
   fildyn='Cu.dyn',
  /

Output Gamma:

Computing dynamical matrix for.
                     q = (   0.0000000   0.0000000   0.0000000 )

       7 Sym.Ops. (with q -> -q+G )
//
Atomic displacements:
      There are  42 irreducible representations
//
  Representation #  1 mode #   1

      Self-consistent Calculation

      Pert. #  1: Fermi energy shift (Ry) =    -0.1403E+01     
-0.8470E-21

       iter #   1 total cpu time :   168.5 secs   av.it.:   6.1
       thresh= 0.100E-01 alpha_mix =  0.150 |ddv_scf|^2 =  0.277E+00
//
       Pert. #  1: Fermi energy shift (Ry) =     0.1555E+08     
-0.8882E-15
//
       iter # 100 total cpu time :  2038.3 secs   av.it.:  25.3
       thresh= 0.100E-01 alpha_mix =  0.150 |ddv_scf|^2 =  0.313E+15


And the output for a q-point away from Gamma:

Computing dynamical matrix for.
                     q = (   0.1443376   0.2500000   0.0000000 )

       2 Sym.Ops. (no q -> -q+G )
//
  Atomic displacements:
      There are  63 irreducible representations
//
Representation #  1 mode #   1

      Self-consistent Calculation

       iter #   1 total cpu time :  1117.1 secs   av.it.:   5.8
       thresh= 0.100E-01 alpha_mix =  0.150 |ddv_scf|^2 =  0.315E-04
//
       iter #  15 total cpu time :  2442.8 secs   av.it.:  10.2
       thresh= 0.105E-06 alpha_mix =  0.150 |ddv_scf|^2 =  0.454E-12



Some info on the self-consistent calculation before:

occupations='smearing', smearing='methfessel-paxton', degauss=0.01
ecutwfc = 38.0, ecutrho = 304.0,
conv_thr =  1.0d-10
K_POINTS {automatic}: 12 12 1 0 0 0
nat= 21, ntyp= 1
CELL_PARAMETERS {hexagonal}
0.86602540    -0.5000000   0.000000
0.86602540    0.50000000   0.000000
0.00000000    0.00000000   25.71450



-------------------------------------------------------------
Nicki Frank Hinsche, Dr. rer. nat.
Institute of physics - Theoretical physics,
Martin-Luther-University Halle-Wittenberg,
Von-Seckendorff-Platz 1, Room 1.07
D-06120 Halle/Saale, Germany
Tel.: ++49 345 5525460
-------------------------------------------------------------

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20131027/dfe7295d/attachment.html>